Advantest R6581 8.5 digit DMM mini teardown/repair

[leighcorrigall]

Shematics R6581 v.5

Great. Even though I had Mickle's zipped folder, I didn't know this existed. Thank you for pointing this out.

[serg-el]

Great. Even though I had Mickle's zipped folder, I didn't know this existed. Thank you for pointing this out.

Not the last one.  There are also minor bugs.  And far from everything has been tested and drawn.
Together with Mickle we are working on this.  Thanks a lot to him

[leighcorrigall]

Since there are no translations from Japanese that I am aware of for the Advantest R6581, attached is my own drafted translation for the Calibration section (with the exception of the 'ADC Commands'). I plan on translating the rest of the manual in steps, based on my own needs. I hope this helps our community. Let me know if there are corrections to be made when you find them. The file is zipped (7Z) because this website does not accept DOCX for whatever reason.

Interestingly enough, I discovered that the instrument has three internal temperature probes for the DCV, OHM, and AC regions (see 13.6.1 SCPI Commands). It is a nice feature of the remote commands.

どうもありがとうございました

[LaserEng]

I have a R6581 on the way to me with damaged front banana jacks and no display, I haven't read the full thread yet but a few questions regarding these meters.

Are the front and rear banana jacks interchangeable ?

Does anyone know what the model number of VFD is so I can research if it's still available, I'm guessing it will most likely be Noritake Itron VFD though I'm hoping the blank display is just power supply trouble.

[picburner]

Someone here has the original new display for sale but at a high price.

[leighcorrigall]

I have a R6581 on the way to me with damaged front banana jacks and no display, I haven't read the full thread yet but a few questions regarding these meters.

Are the front and rear banana jacks interchangeable ?

Does anyone know what the model number of VFD is so I can research if it's still available, I'm guessing it will most likely be Noritake Itron VFD though I'm hoping the blank display is just power supply trouble.

Do not buy from the eBay seller; it is triple the price. Talk to Pipelie about a new screen. He charges a reasonable price and still has a few spares. I just received one from him and it is identical to the original.



Talk to Branadic about the binding post adaptors. There are some posts about it on the EEVBlog if you search for them.

Good luck!

[branadic]

Are the front and rear banana jacks interchangeable ?

Yes they are, but you can't score the original banana jacks anywhere. As some of mine were broken too, I replaced them with low thermal binding posts. Therefore an adapter was needed, that is the same for front and rear, read here for more and to find the CAD files:

https://www.eevblog.com/forum/metrology/advantest-r6581-8-5-digit-dmm-mini-teardownrepair/msg2663463/#msg2663463

I also bought a display from pipelie, the replacement was an easy task.

-branadic-

[leighcorrigall]

Talk to Pipelie about a new screen. He charges a reasonable price and still has a few spares. I just received one from him and it is identical to the original.

Attached are some photos of the replacement screen I installed tonight. Pretty decent if you ask me.

Replacement suggestions:
1] Heat a thin metal blade with a hot gun and carefully insert it between the VFD and the adhesive tape to loosen the bond.
2] Pour isopropyl alcohol into the crack with a Q-tip or cloth and wait until the adhesive dissolves.
3] Continue this process until the display is separated from the board.
4] Completely desolder the VFD leads. Check that each lead is desoldered before removing the display. You do not want to damage the board through-hole contacts.
5] Clean the board with isopropyl alcohol.
6] Use a sewing needle of the same size as the VFD through-holes to ensure that the new VFD leads are free to be inserted. If you find a hole that is too small, heat the needle while inserting it into the hole with a soldering iron. Never force the VFD pins into the holes or you will damage them and possibly the display.
7] Lay the VFD on the side with the pins and roll them slightly so that all the pins are evenly bent by the table's surface. Never over bend the pins or you could break them off. Check that each pin is evenly spaced and perpendicular to the VFD. To seat the pins, start on one side and wiggle them gently into place. This requires a lot of care and attention. Use tweezers to guide any stray pins.
8] Once the display pins have been seated properly, solder the pins on the sides so that the VFD cannot be removed again. Lift the VFD slightly to add an adhesive between the VFD and the cushion. I used an epoxy with a syringe.
9] Solder the remaining pins. Let the adhesive dry enough to continue working.
10] Check for shorting with a handheld DMM before reconnecting the display and turning on the instrument. I checked for shorting three times, just to be certain. You might also want to check to ensure all soldering has been done well too at this time.
11] Hold the HOME key and power ON the Advantest R6581. Select the DISPLAY option to check that the display has been properly installed.
12] If the display is working correctly, have a beer. Otherwise, diagnose the VFD problem. 



Thank you Ian J. for your video tips. https://youtu.be/wXQdo42C6pE?t=351 They came in handy, but I had to improvise because the R6581 is much more difficult to install.

[leighcorrigall]

And just for completeness I received two 3D printed adapters with a ceramic filled resin today, so I changed the front adapter and also installed the rear binding posts. There is no drilling or other modification needed, just simply install the binding posts to the adapter and screw it down to the chassis.

-branadic-

Hi Branadic,

Would you please explain how to connect the binding posts to the mainboard? I intend to perform this modification also.

Is it reasonable to replace the original crimp connectors with quality copper spade connectors or is soldering involved?

Thanks.

[branadic]

@ leighcorrigall

Congrats to your new display. As far as I can tell it is much brighter than the VFD on 3458A.
I removed the original tinned crimp contacts and used some copper hook lugs, that I got from a friend. But you could also solder the PTFE wires directly to the binding posts, no problem.

-branadic-

[leighcorrigall]

Here is my experience installing the LowThermal binding posts and adaptor:

1] The soldering and insulation of the terminations were relatively simple along with mounting the adaptor to the aluminum panelling. Make sure to solder the wires according to your hand dominance to avoid accidentally melting something with the soldering iron. I used a temperature of about 850  °C so that the solder melted quickly and the posts would only be heated locally. The solder I used contained 40 % Pb, which made it much easier to melt and bond to the copper post.


2] I was upset when I discovered that the LowThermals might not be compatible with every R6581 model. At this point, I already cut the wires from the original banana terminals. On further inspection, I realized that there was a misalignment with the front panel assembly. In my case, the plastic panel label covering was misaligned with the aluminum back panel. I first attempted to unscrew the analogue board case from the rest of the instrument in an attempt to align the parts. This did not improve the situation.


3] I ended up dismounting the front panel and using a razor blade to remove access plastic causing the misalignment. To do this properly, you must remove a small portion of the material at a time or it will result in jagged edges instead of a smooth circular shape for the hole. Luckily, I noticed this quickly enough that I was able to correct the error before I ran out of material. Note that the disproportionate line thicknesses of the coloured circles around each through-hole will not be noticeable once the binding posts are installed. I suppose another strategy would be to delaminate the plastic from the aluminum, but you might end up misaligning something else and you risk creasing the plastic. I took a shortcut.


4] Here is a picture of the backside of the front panel. There was enough wire that I was able to wind the exposed wire around each binding post at least once before soldering. Wrapping and soldering will provide a decent connection. Make sure to inspect the soldering of each job or you might end up with problems later on.


5]
The binding posts were tested by measuring a 10 Ω Vishay resistor in 4-wire resistance mode. Just to be clear, the front panel colours on my Advantest R6581T were blue, green, yellow for the top and orange, white, red for the bottom row. My DMM is due for calibration, so I am not concerned about any modifications at this point. I have plenty of calibrated testing equipment as a substitute for this 8.5 digit DMM.

[leighcorrigall]

Incorrect scheme, from the manufacturer.
Expand the capacitor.



  Wrong.



  Circuit diagram.



  Right.

Is the C215 in the attached photo also incorrectly mounted? The silkscreen says that it is correctly oriented, but I suppose it might not follow the circuit exactly. The calibration dates shown on my Advantest are for 2005/08/01 to 2015/07/31, making me believe it to be a newer variation. The serial number is 56230037. Was the mistake made in later models?

[Kleinstein]

If in doubt, one could just measure the voltage at the cap. They may survive the wrong polarity for some time.

[leighcorrigall]

If in doubt, one could just measure the voltage at the cap. They may survive the wrong polarity for some time.

The board has been removed to upgrade R200 and R234 among other changes. I stumbled upon this and I figured I might as well consider this as well since the board is already out.

[leighcorrigall]

That is some mistake. I suppose I will reverse the capacitor as I am reading -5.5 DCV at the positive side. 

[serg-el]

Have you altered the control of the K006 relay?  This must be done. 
But I do not see the need to replace the resistors.

[leighcorrigall]

Have you altered the control of the K006 relay?  This must be done. 
But I do not see the need to replace the resistors.

Yes, I did. Thank you for asking! 

Now all I have to do is have this instrument calibrated. It is a shame that I do not have a calibrated precision 10 DCV source yet. One thing at a time.

[leighcorrigall]

The calibration instructions for this instrument recommend < 12 AWG wire as a short, but this seems to be a bit excessive and may actually be less effective. I used a spent OFHC copper (C10100) Conflat-40 gasket from a turbomolecular vacuum system, a Keysight 11059A Kelvin Probe Set and tin-foil shielding for now. 

These are the shorts that I have found available:
Fluke 884X-Short (70 CAD)
Keithley 8610 (110 CAD)
Keysight 34172B (290 CAD)

Does anyone have recommendations for a low-thermal 4-wire calibration short? Is there anything better than my silly substitute?

EDIT: I ended up purchasing a Keithley 8610 because it stated that it was a low-termal short in the datasheet and was listed on the Keithley 2002 (8.5-digit DMM) datasheet list of accessories.

[Kleinstein]

The short does not really need very thick copper. The one for the DMMs are usually just a PCB with 4 connectors.

There are plans for a makeshift 4 wire short by just bending a piece of some 1-1.5 mm² bare copper wire to make contact inside the 4 mm connectors. With the screw terminals one could also just use a thinner copper wire in a kind of U shape.

[branadic]

Question to all R6581 owners, would you like to contribute translating the instruction manual? I already started with the index, but to be honest, there are a lot of pages. But if everyone contributes only a little piece, we would then be able to put things together faster and could have a full manual soon.

I do have the japanese OCR pdf-file transfered into an .odt file and translated, as good as google translate could do the job based on the signs and letters that were detected by the OCR program. Unfortunately sometimes the OCR program failed to detect the proper sign/letter. So there is still a lot of work to do, checking and verifying, putting pictures and tables into a proper state. Anyone interested to support that work?

-branadic-

[leighcorrigall]

Question to all R6581 owners, would you like to contribute translating the instruction manual? I already started with the index, but to be honest, there are a lot of pages. But if everyone contributes only a little piece, we would then be able to put things together faster and could have a full manual soon.

I do have the japanese OCR pdf-file transfered into an .odt file and translated, as good as google translate could do the job based on the signs and letters that were detected by the OCR program. Unfortunately sometimes the OCR program failed to detect the proper sign/letter. So there is still a lot of work to do, checking and verifying, putting pictures and tables into a proper state. Anyone interested to support that work?

-branadic-

I second that motion.

I have already translated most of the calibration section apart from a communication example. It should be attached to this thread as a DOCX within a compressed file.

... attached is my own drafted translation for the Calibration section (with the exception of the 'ADC Commands')...

May I suggest that we provide credit to each translation contribution that is incorporated into the English version? Something as simple as the author's name in the header would be much appreciated.

Also, once we have enough contributors organized, we should assign a reviewer to read the entire document to check for consistency and errors. This could be more than one person. A revision number can be included too.

My other suggestion would be to translate the GPIB handbook as a lot of members would like to take advantage of this resource.

Regards.

[serg-el]

Since there is no ACV calibration video on the network, I decided to fix it.

https://youtu.be/uMjWAdwoWYI

[Mickle T.]

Advantest R6581 RMS board is like a work of art. I found and google-translated one patent that sheds light on the ACV self-calibration technique. But this is just a guess. Real equipment is much more complex and interesting. Therefore, reverse engineering of the RMS board is on my to-do list (if, of course, one day I have it).

[branadic]

Here are the first few pages, chapter 13 and the index merged together.
I've added a page with the contributor.

-branadic-

[gamalot]

Here are the first few pages, chapter 13 and the index merged together.
I've added a page with the contributor.

-branadic-

I took a quick glance at the translation of Chapter 13 and found that it is different from the manual I downloaded from xDevs and KO44B. There is no Figure 1 in my version.

I also found the Teflon test lead mentioned in 13.1 should be "high-impedance" or "high insulation resistance", not "low-impedance". In addition, Figure 6 is about "ZERO FRONT", but the picture is obviously the back panel.

I tried to translate the Japanese in Figure 1 into English, and I attached it.